Golden Samples and Retained Samples: How Spice Buyers Keep Every Lot Consistent (2026)
Short answer: a golden sample is the sealed, approved physical reference for your spice or blend, and a retained sample is a sealed portion kept from every lot you ship. Buyers who keep every lot consistent make the golden sample from an approved pilot lot, store it the same way on both sides, compare every pre shipment sample against it before release, refresh it on a schedule, and keep retained samples long enough to settle any complaint with a test instead of an argument.
This guide is for importers, retailers, private label brands and foodservice buyers who buy Indian spices and masala blends in repeat lots. It explains the sample types used in the spice trade, how to create, label and store a golden sample, how to compare a lot against it, when to refresh it, and how retained samples protect you when a customer says the product has changed.
Who this is for: buyers who have approved a spice or blend once and now need the tenth, fiftieth and hundredth lot to look, smell and taste the same on the shelf, in the restaurant kitchen or on the production line.
In this guide
Why spice lots drift even when the recipe does not change
Spices are agricultural products. The same formula made from a different harvest, a different growing region or a different storage age can give a blend that looks, smells or tastes noticeably different, even when every number on the specification sheet still passes.
The common causes are easy to list and hard to control without a physical reference:
Crop year and origin. Colour, heat and volatile oil vary with weather, variety and growing region, so a new crop can shift a blend even at the same proportions.
Age of the raw spice. Aroma fades with time in storage, so a lot made from older stock can be flatter than one made from fresh material.
Grinding and handling. Grind size and the heat generated during grinding change how quickly aroma is released and lost.
Moisture and packing. Small differences in moisture or packaging barrier change how a blend flows, clumps and ages.
A specification sheet describes the product in words and numbers. It cannot fully describe aroma, the balance between spices or the impression a blend gives in a finished dish. That is the job of the golden sample, and it is why buyers who rely on numbers alone often notice drift only after their customers do.
Key point: numbers tell you a lot is safe and within limits. The golden sample tells you it is still your product. You need both to keep every lot consistent.
The sample types every spice buyer should know
Suppliers and buyers use several names for samples, and confusion between them causes real disputes. Agree the names and what each one is for at the start.
Sample | Where it comes from | What it is for | Who keeps it |
|---|---|---|---|
Reference or target sample | The buyer's current product or a competitor product | Showing the flavour target during development | Buyer, with a portion sent to the supplier |
Development sample | Trial versions made during development | Choosing the direction of the formula | Both, until a version is approved |
Golden sample | The approved version, ideally from an approved pilot lot | The physical standard every future lot is compared with | Both, sealed and labelled |
Type sample | A representative sample of a grade of whole spice | Showing the grade offered in bulk trade | Both, for the life of the contract |
Pre shipment sample | Drawn from the actual production lot before dispatch | Approving the lot before it ships | Buyer, with a matching portion kept by the supplier |
Retained sample | Sealed portion of every shipped lot | Investigating complaints and retesting | Both, for the agreed retention period |
Arbitration sample | An extra sealed portion of the lot | Testing at an agreed independent lab if results are disputed | Held sealed, opened only for a dispute |
Some suppliers call the golden sample a master sample or approved standard, and some call retained samples counter samples or keep samples. The name matters less than writing down which physical sample is the standard and where each portion is kept.
How to create a golden sample
A good golden sample represents what the supplier can actually make at scale, not the best thing a lab bench can produce once. Get this step right and most later disputes never happen.
Start from an approved pilot lot
Make the golden sample from a pilot or first production run that you have approved, rather than from a small hand made development sample. A blend mixed in a few hundred grams can behave differently from the same formula ground and mixed at commercial scale. At Harvestia, pilot runs start from 50 kg, which gives a realistic lot to approve and to draw the golden sample from.
Approve it in your application
Approve the blend the way your customers will meet it: in the curry, on the snack, in the restaurant dish or straight from the shaker. Write down how you tested it, at what dosage and in which base recipe, so that every later comparison is made the same way.
Split, seal and label it
Divide the approved material into identical portions at the same time, from the same mixed lot, so that every portion starts the same. Seal each portion in the same packaging and keep at least one portion on each side. Label every portion with:
Product name and the formula or specification version code.
The lot number the golden sample was drawn from.
Approval date and the names of the people who approved it on both sides.
The planned refresh or review date.
The storage condition it must be kept in.
Link the golden sample to your specification sheet by that version code. Our spice blend specification sheet template shows where the golden sample reference sits alongside the measured limits.
Key point: if the golden sample and the specification sheet carry different version codes, nobody can say which standard a lot was made to. One version code should tie the spec, the golden sample and every lot code together.
Storing golden and retained samples so they stay useful
A golden sample that has lost its aroma, picked up moisture or absorbed odours from the shelf next to it is no longer a fair standard. Storage is where most sample programs quietly fail.
Packaging
Use packaging that protects against air, light and moisture: airtight glass jars or high barrier laminate pouches, filled with as little headspace as practical. Avoid thin clear plastic bags, which let in light and air and can transfer odour. Use the same packaging for every portion, so that the buyer's and the supplier's portions age the same way.
Storage conditions
Keep samples cool, dry, dark and away from strong odours, at a stable temperature. Agree the condition in writing and store the portions under the same condition on both sides; a golden sample kept in a warm office on one side and a cool store on the other will drift apart and make every comparison unreliable.
How much to keep
Keep enough material to repeat every test on your specification and run a sensory comparison, with some left over for a second round if the first result is challenged. Agree the quantity in grams in your specification, because the right amount depends on your test list and on how many people take part in sensory checks.
How long to keep retained samples
Keep retained samples at least for the shelf life of the lot, plus the period in which a customer complaint or a regulatory question could realistically arrive. Write the retention period into your specification or supply agreement so that both sides keep samples for the same length of time.
Refreshing the golden sample without losing your taste
Even well stored spices slowly lose aroma and colour, so a golden sample cannot be kept forever. If it is never refreshed, a fresh, correct lot can start to look "too strong" against a tired standard, and good lots get rejected.
Set a refresh schedule
Agree a refresh interval based on the shelf life of the product and how quickly its aroma fades, and put the next review date on the label. Spices with delicate volatile aromas usually need more frequent refreshing than robust whole seeds.
Bridge the old standard to the new one
When you refresh, do not simply swap in a new sample. Compare the candidate for the new golden sample against the current golden sample and against retained samples from recent lots that customers were happy with. Approve the new standard only when it matches the product your customers know, then retire the old one with a note in the change log.
Triggers for a new golden sample
Trigger | What changes | What to do |
|---|---|---|
Scheduled refresh date | Aroma and colour of the stored standard fade | Make a new portion from a recent approved lot and bridge it to the old one |
New crop year | Colour, heat or aroma of raw spices shift | Run a pilot lot, compare with the current standard and approve before full production |
Change of origin or grade | Component character changes | Treat it as a new version: new spec version, new golden sample |
Formula change | Proportions or ingredients change | Issue a new spec version code and a new golden sample |
Storage failure | Standard damaged, opened or contaminated | Replace from a retained sample of a well received lot and record why |
Using the golden sample to release a lot
The golden sample earns its keep at lot release. A simple, repeatable release routine catches drift before the goods leave India, which is far cheaper than finding it at your warehouse.
Draw a representative pre shipment sample. Take it across the lot, from several bags or cartons, not one scoop from the top of one bag. ISO 948 sets out sampling practice for spices and condiments and is a useful reference for writing your sampling rule.
Compare it side by side with the golden sample. Use the same conditions every time: same light, same quantity, same base recipe, same people where possible.
Check the measured parameters. Read the lab report or COA for the lot against your specification limits, and look at any parameter that is drifting towards a limit.
Record the decision. Note who approved the lot, the date, the comparison result and any comments.
Seal the retained samples. Seal portions of the same pre shipment material for both sides, labelled with the lot number, before the lot ships.
If you are new to reading lab reports, our guide to what global buyers check before sourcing from Indian spice exporters covers the parameters most importers ask for.
How to compare a lot against the golden sample
A comparison is only useful if it is fair and repeatable. Decide in advance what you compare, how you compare it and what counts as a match.
Appearance and colour
Compare equal amounts side by side on a white surface under the same light, ideally daylight balanced light rather than mixed room lighting. Where colour matters commercially, add a measured value: extractable colour for chilli and paprika, curcumin content for turmeric, or instrumental colour readings if both sides have compatible equipment.
Aroma and taste
Smell and taste the lot and the golden sample in your application, at the agreed dosage, with the samples coded so the tasters do not know which is which. A triangle test, where tasters receive three coded samples and pick the odd one out, is a simple, recognised way to check whether a difference is real; it is described in ISO 4120. If tasters cannot reliably pick the lot out from two portions of the golden sample, you have a practical match.
Measured parameters
Use the numbers that explain what you sense. Moisture affects flow and ageing, volatile oil (ISO 6571) tracks aroma strength, particle size affects texture and release, and heat or colour markers explain visual and pungency differences.
What you compare | How | Typical evidence |
|---|---|---|
Colour | Side by side under the same light, plus a measured value where it matters | Colour reading, curcumin or extractable colour result |
Aroma and taste | Coded samples in your application, triangle test for close calls | Sensory record with tasters, dosage and result |
Heat | Tasting in application, plus a measured pungency value where agreed | Lab result against your limit |
Texture and grind | Side by side visual check and particle size result | Sieve analysis result |
Moisture and volatile oil | Lab results on the lot | COA lines against your specification |
Key point: agree what counts as a match before the first lot. "Comparable to golden sample version 3 in the agreed base recipe at the agreed dosage" is a standard both sides can test. "Should taste the same" is not.
When a customer complains: using retained samples
Retained samples turn a complaint into an investigation. Without them, the buyer has only the customer's description and the supplier has only its records, and both sides end up defending positions instead of checking the product.
A simple investigation routine
Identify the lot. Use the lot code on the pack to find the lot number, the release record and the COA.
Pull the retained samples on both sides. Check that the seals are intact and the samples were stored as agreed.
Compare the retained sample with the golden sample. If the retained sample matches and the complaint sample does not, the issue probably arose after shipment, in storage, distribution or use.
Test where needed. If results are disputed, open the arbitration sample at an agreed accredited lab; ISO/IEC 17025 accreditation for the relevant tests is a sensible requirement.
Decide and record. Agree the outcome, the corrective action and whether the specification, the golden sample or the release routine needs to change.
Write the rules down before you need them
Your supply agreement or specification should say who keeps which samples, for how long, which lab settles disputes, who pays for the test and how quickly each side must respond. These are commercial terms to agree with your supplier and your own advisers; they are much easier to settle on a calm day than during a complaint.
Common sample management mistakes
Mistake | What goes wrong | Fix |
|---|---|---|
Golden sample made from a bench sample | Commercial lots never quite match it | Draw the golden sample from an approved pilot or first lot |
Only one side keeps the golden sample | No shared standard when results differ | Split identical portions and keep at least one on each side |
Different storage on each side | The two standards drift apart | Same packaging and the same agreed storage condition |
No refresh date | Fresh, correct lots fail against a faded standard | Put a review date on the label and bridge each refresh |
Retained samples not sealed or labelled | They cannot be trusted in a dispute | Seal and label by lot number before the lot ships |
Retention period shorter than shelf life | Samples are gone when a complaint arrives | Keep retained samples for shelf life plus a claims margin |
Comparison method not written down | Every lot is judged differently | Fix the application, dosage, light and test in the specification |
Buyer checklist
Use this list when you set up a new spice or blend, and review it at every golden sample refresh.
Agree the names of each sample type and what each one is used for.
Make the golden sample from an approved pilot or first commercial lot.
Approve the golden sample in your own application and record the base recipe and dosage.
Split identical portions at the same time and keep at least one sealed portion on each side.
Label every portion with product, version code, source lot, approval date, approvers, refresh date and storage condition.
Use the same packaging and the same agreed storage condition on both sides.
Write the comparison method and what counts as a match into the specification.
Ask for a representative pre shipment sample from every lot and compare it with the golden sample before release.
Read the COA for every lot against your specification limits.
Seal and label retained samples of every lot on both sides before shipment.
Set the retention period at least equal to shelf life plus a realistic claims margin.
Agree the arbitration lab, test costs and response times for disputes in advance.
Refresh the golden sample on schedule and bridge each new version to the old one.
How Harvestia works with golden and retained samples
Harvestia Group works with importers, retailers, distributors, food brands and foodservice buyers who need the same product in every lot. The way we work follows the order that protects consistency:
Source: spices come through a clear supply chain from ground level farming families, with about 8,000 farming members across India helping maintain quality.
Quality control: incoming checks on raw spices, comparison against your approved golden sample, and lab reports / COA with commercial lots against the parameters in your specification.
Pack under your brand: we process and supply, and we pack under the buyer's brand in the format you approve.
FOB Mundra: export lots are prepared for shipment FOB Mundra, with documents coordinated with your forwarder.
We agree the golden sample, the comparison method and the retained sample rules with you before the first commercial lot, so that both sides are checking the same standard in the same way. Samples are available from Rs 250 and pilot runs start from 50 kg, which gives you a real lot to approve and to draw your golden sample from. If you are building a signature blend, our custom masala manufacturer guide explains how the brief, the golden sample and the specification fit together.
Prefer email? Write to sales@harvestiagroup.com with the spices or blends you need, your destination market, your channel and estimated volumes, and send a reference sample if you have one.
FAQs
What is a golden sample in the spice trade?
A golden sample is the sealed, approved physical reference for a spice or blend. Buyer and supplier each keep an identical portion, and every future lot is compared against it for appearance, aroma, taste and other agreed parameters.
What is the difference between a golden sample and a retained sample?
A golden sample is the standard every lot is compared with; a retained sample is a sealed portion of one specific lot. You keep one golden sample per product version and one retained sample for every lot shipped.
Is a golden sample the same as a pre shipment sample?
No. A pre shipment sample is drawn from the actual lot before dispatch and is compared with the golden sample to decide whether that lot can ship.
How much golden sample should each side keep?
Each side should keep enough to repeat every test on the specification and run a sensory comparison, with some left for a second round. Agree the quantity in grams in your specification, because it depends on your test list.
How long should retained samples be kept?
Keep retained samples at least for the shelf life of the lot plus a realistic period for complaints to arrive. Write the period into your specification or supply agreement so both sides keep them equally long.
How often should a golden sample be refreshed?
Refresh it on a schedule tied to the product's shelf life and how fast its aroma fades, and also whenever the crop year, origin, grade or formula changes. Bridge each new standard to the old one using retained samples from well received lots.
How do you compare a spice lot with the golden sample fairly?
Compare equal amounts side by side under the same light and in your application at the agreed dosage, with coded samples. Back the sensory result with measured values such as moisture, volatile oil, colour and particle size.
What is a triangle test for spices?
A triangle test gives tasters three coded samples, two the same and one different, and asks them to pick the odd one out. It shows whether a difference between a lot and the golden sample is real, and it is described in ISO 4120.
Who should hold the arbitration sample?
The arbitration sample should stay sealed and be opened only for a dispute, at a lab both sides agreed in advance. Where it is held and who pays for testing are commercial terms to settle in your supply agreement.
Can Harvestia work with my existing golden sample?
Yes. Send your reference or golden sample with your specification, and we will match against it, starting with samples from Rs 250 and a pilot run from 50 kg that you approve before the first commercial lot.
Want every lot to match the one you approved?
Send us your product list, your destination market, your channel and a reference sample if you have one. We will come back with questions, a sample plan and a quote.
Disclaimer: this article is commercial guidance only. It is not legal, customs or regulatory advice. Standards and method references are examples; always confirm current requirements for your products and markets with the relevant authorities and your own advisers.
Harvestia Group, Ahmedabad, Gujarat, India | sales@harvestiagroup.com | www.harvestiagroup.com
Comments